考虑人类视觉特征的融合图像评价方法
|
摘要
图像质量评价是遥感图像融合必不可少的环节,对融合图像的客观评价具有较好的研究价值,好的客观评价指标应尽量与主观评价一致。针对全色图像和多光谱图像的融合质量评价,在考虑人类视觉系统的多通道分解和对比度敏感特性的基础上,首先修改颜色失真和结构相似度来分别评价光谱信息和空间信息,然后综合两者作为融合图像的整体评价。以SPOT5图像和高分二号卫星图像为原始数据,对7种融合图像的质量进行了评价。实验表明,对于高分辨率的遥感图像,这种引入了人类视觉特征的融合评价方法与主观评价有很好的一致性。
Image quality assessment plays an important role in remote sensing image fusion. Research on objective evaluation methods is of great value in this field. Because the fused images will be observed by human finally, the best objective evaluation index of image quality should be consistent with the subjective evaluation. After the human visual characteristics are introduced into, image quality evaluation method can get better result. Therefore, based on the multi channels decomposition and contrast sensitivity characteristics of human visual system, this paper proposes a new method to evaluate the fusion quality of panchromatic image and multispectral image. Firstly, the color distortion and structural similarity are modified to evaluate the spectral information and spatial information respectively. Secondly, they are integrated as the overall evaluation of fused image. According to the experiment evaluating quality of different fused images by SPOT image data and GF-2 image data, it proves that the proposed method is in good agreement with the subjective evaluation when the remote sensing image is high resolution data.
Image quality assessment plays an important role in remote sensing image fusion. Research on objective evaluation methods is of great value in this field. Because the fused images will be observed by human finally, the best objective evaluation index of image quality should be consistent with the subjective evaluation. After the human visual characteristics are introduced into, image quality evaluation method can get better result. Therefore, based on the multi channels decomposition and contrast sensitivity characteristics of human visual system, this paper proposes a new method to evaluate the fusion quality of panchromatic image and multispectral image. Firstly, the color distortion and structural similarity are modified to evaluate the spectral information and spatial information respectively. Secondly, they are integrated as the overall evaluation of fused image. According to the experiment evaluating quality of different fused images by SPOT image data and GF-2 image data, it proves that the proposed method is in good agreement with the subjective evaluation when the remote sensing image is high resolution data.
引文
[1] PohI C, van Gend'eren J L. Multisensor Image Fusion in Remote Sensing: Concepts, Methods and Application [J]. International Journal of Remote Sensing, 1998, 19(5): 823-854
[2] Han Ling. Study of Multi-source Remote Sensing Image Fusion Technology and Multi-source Information’s Application to Geoscience[D]. Xi’an: Northwest University, 2005(韩玲. 多源遥感信息融合技术及多源信息在地学中的应用研究[D]. 西安:西北大学, 2005)
[3] Ma Xudong. Research on Quality Assessment of Image Compression and Image Fusion for Optical Remote Sensing Image[D]. Wuhan: Wuhan University, 2014(马旭东. 光学遥感影像压缩及融合的质量评价研究[D]. 武汉:武汉大学, 2014)
[4] Li Yanfang, Cheng Xinwen, Jin Biao, et al. Comparison of Several Common Image Fusion Methods[J]. Geospatial Information, 2009, 7(4): 102-105(李艳芳, 程新文, 金彪,等. 几种常见影像融合方法的比较[J]. 地理信息系统, 2009, 7(4):102-105)
[5] Maria G A, Jose L S, Raguel G C, et al. Fusion of Multispectral and Panchromatic Images Using Improved HIS and PCA Mergers Based on Wavelet Decomposition[J]. IEEE Transactions on Geoscience and Remote Sensing, 2002, 42(6): 1 291-1 299
[6] Xing Shuai. Research on Image Registration and Image Fusion for Multi-sensors Remote Sensing Image[D]. Zhengzhou: Information Engineering University, 2004(邢帅. 多源遥感影像配准与融合技术的研究[D]. 郑州: 信息工程大学, 2004)
[7] Wang Zhijun, Li Deren, Li Qingquan. Fusion of SPOT-P and TM Multispectral Image by M-band Wavelet Theory[J]. Geomatics and Information Science of Wuhan University, 2001, 26(1): 24-28(王智均, 李德仁, 李清泉. 多进制小波理论在SPOT和TM影像融合中的应用[J]. 武汉大学学报·信息科学版, 2001, 26(1):24-28)
[8] Yan Li, Yue Xijuan, Cui Chenfeng. A Method to Quantitatively Determine Spatial Resolution of RS Fused Image[J]. Geomatics and Information Science of Wuhan University, 2007, 32(8): 667-670(闫利, 岳昔娟, 崔晨风. 一种定量确定遥感融合图像空间分辨率的方法[J]. 武汉大学学报·信息科学版, 2007, 32(8):667-670)
[9] Daly S J. The Visible Difference Predictor: An Algorithm for the Assessment of Image Fidelity[J]. Digital Image and Human Vision, 1993, 1 666: 179-206
[10] Lubin J. A Visual Discrimination Model for Image System Design and Evaluation[OL]. https://doi.org/10.1142/9789812831200_0010,1995
[11] Wang Z, Bovik A C, Sheikh H R. et al. Image Quality Assessment: From Error Visibility to Structural Similarity[J]. IEEE Trans Image Processing, 2004, 13(4): 600-612
[12] Wang Z, Bovik A C, Lu L. Why Is Image Quality Assessment so Difficult [C]. IEEE Int Conf Acoust Speech and Signal Processing, Orlando, FL, USA, 2002
[13] Aja-Fernandez S, Estepar R S J, Westin C F, et al. Image Quality Assessment Based on Local Variance[C]. The 28th IEEE EMBS Annual International Conference, New York, USA, 2006
[14] Ma Xudong, Yan Li, Cao Wei, et al. A New Image Quality Assessment Mode Based on the Gradient Information[J]. Geomatics and Information Science of Wuhan University, 2014, 39(12): 1 412-1 418(马旭东, 闫利, 曹纬, 等. 一种新的利用梯度信息的图像质量评价模型[J]. 武汉大学学报·信息科学版, 2014, 39(12):1 412-1 418)
[15] Liu Jun, Shao Zhenfeng. Feature-Based Remote Sensing Image Fusion Quality Metrics Using Structure Similarity[J]. Acta Photonica Sinica, 2011, 40(1): 126-131(刘军, 邵振峰. 基于特征结构相似度的遥感影像融合质量评价指标[J]. 光子学报, 2011, 40(1): 126-131)
[16] Liu Shuqin, Wu Lifang, Gong Yu, et al. Overview of Image Quality Assessment[J]. Sciencepaper Online, 2011, 6(7): 501-506(刘书琴, 毋立芳, 宫玉, 等. 图像质量评价综述[J].中国科技论文在线, 2011, 6(7): 501-506)
[17] Carper W J. The Use of Intensity-Huesaturation Transformations for Merging SPOT Panchromatic and Multispectral Image Data[J]. Photogrammetric Engineering and Remote Sensing, 1990, 56(4): 459-467
[18] Liu Yuqiong. Study on Pixel-Level Multi-source Remote Sensing Data Fusion Technique[D]. Xi’an: Chang’an University, 2007(刘宇琼. 基于像素级多源遥感影像融合技术的研究[D]. 西安:长安大学, 2007)
[19] Mallat S G. A Theory for Multi-resolution Signal Decomposition: The Wavelet Representation[J]. IEEE Transation PAMI, 1989, 11(7): 674-693
[20] Wang Zhengyou, Huang Longhua. Image Quality Assessment Method Based on Contrast Sensitivity[J]. Computer Application, 2006, 26(8): 1 857-1 859(王正友, 黄隆华. 基于对比度敏感度的图像质量评价方法[J]. 计算机应用, 2006, 26(8): 1 857-1 859)
[21] Shen J H. On the Foundations of Vision Modeling I. Weber’s Law and Weberized TV Restoration[J]. Physica D Nonlinear Phenomena, 2003, 175(3): 241-251
[22] Mannos J L, Sakrison D J. The Effects of a Visual Fidelity Criterion on the Encoding of Images[J]. IEEE Trans Information Theory, 1974, 4:525-536
[23] Hu Xuming, Zhang Dengfu, Nan Dong, et al. Method of Image Visual Quality Evaluation Based on Human Visual Characteristics[J]. Computer Application, 2012, 32(7): 1 882- 1 884,1 889(胡许明, 张登福, 南栋,等. 基于人眼视觉特征的图像视觉质量评价方法[J]. 计算机应用, 2012, 32(7):1 882-1 884, 1 889)
[24] Li Wenjuan. Study on the Algorithms for HVS-Based Color Image Measurement[D]. Changchun: Jilin University, 2006(李文娟. 基于视觉特性的彩色图像质量评估研究[D]. 长春:吉林大学, 2006)
[2] Han Ling. Study of Multi-source Remote Sensing Image Fusion Technology and Multi-source Information’s Application to Geoscience[D]. Xi’an: Northwest University, 2005(韩玲. 多源遥感信息融合技术及多源信息在地学中的应用研究[D]. 西安:西北大学, 2005)
[3] Ma Xudong. Research on Quality Assessment of Image Compression and Image Fusion for Optical Remote Sensing Image[D]. Wuhan: Wuhan University, 2014(马旭东. 光学遥感影像压缩及融合的质量评价研究[D]. 武汉:武汉大学, 2014)
[4] Li Yanfang, Cheng Xinwen, Jin Biao, et al. Comparison of Several Common Image Fusion Methods[J]. Geospatial Information, 2009, 7(4): 102-105(李艳芳, 程新文, 金彪,等. 几种常见影像融合方法的比较[J]. 地理信息系统, 2009, 7(4):102-105)
[5] Maria G A, Jose L S, Raguel G C, et al. Fusion of Multispectral and Panchromatic Images Using Improved HIS and PCA Mergers Based on Wavelet Decomposition[J]. IEEE Transactions on Geoscience and Remote Sensing, 2002, 42(6): 1 291-1 299
[6] Xing Shuai. Research on Image Registration and Image Fusion for Multi-sensors Remote Sensing Image[D]. Zhengzhou: Information Engineering University, 2004(邢帅. 多源遥感影像配准与融合技术的研究[D]. 郑州: 信息工程大学, 2004)
[7] Wang Zhijun, Li Deren, Li Qingquan. Fusion of SPOT-P and TM Multispectral Image by M-band Wavelet Theory[J]. Geomatics and Information Science of Wuhan University, 2001, 26(1): 24-28(王智均, 李德仁, 李清泉. 多进制小波理论在SPOT和TM影像融合中的应用[J]. 武汉大学学报·信息科学版, 2001, 26(1):24-28)
[8] Yan Li, Yue Xijuan, Cui Chenfeng. A Method to Quantitatively Determine Spatial Resolution of RS Fused Image[J]. Geomatics and Information Science of Wuhan University, 2007, 32(8): 667-670(闫利, 岳昔娟, 崔晨风. 一种定量确定遥感融合图像空间分辨率的方法[J]. 武汉大学学报·信息科学版, 2007, 32(8):667-670)
[9] Daly S J. The Visible Difference Predictor: An Algorithm for the Assessment of Image Fidelity[J]. Digital Image and Human Vision, 1993, 1 666: 179-206
[10] Lubin J. A Visual Discrimination Model for Image System Design and Evaluation[OL]. https://doi.org/10.1142/9789812831200_0010,1995
[11] Wang Z, Bovik A C, Sheikh H R. et al. Image Quality Assessment: From Error Visibility to Structural Similarity[J]. IEEE Trans Image Processing, 2004, 13(4): 600-612
[12] Wang Z, Bovik A C, Lu L. Why Is Image Quality Assessment so Difficult [C]. IEEE Int Conf Acoust Speech and Signal Processing, Orlando, FL, USA, 2002
[13] Aja-Fernandez S, Estepar R S J, Westin C F, et al. Image Quality Assessment Based on Local Variance[C]. The 28th IEEE EMBS Annual International Conference, New York, USA, 2006
[14] Ma Xudong, Yan Li, Cao Wei, et al. A New Image Quality Assessment Mode Based on the Gradient Information[J]. Geomatics and Information Science of Wuhan University, 2014, 39(12): 1 412-1 418(马旭东, 闫利, 曹纬, 等. 一种新的利用梯度信息的图像质量评价模型[J]. 武汉大学学报·信息科学版, 2014, 39(12):1 412-1 418)
[15] Liu Jun, Shao Zhenfeng. Feature-Based Remote Sensing Image Fusion Quality Metrics Using Structure Similarity[J]. Acta Photonica Sinica, 2011, 40(1): 126-131(刘军, 邵振峰. 基于特征结构相似度的遥感影像融合质量评价指标[J]. 光子学报, 2011, 40(1): 126-131)
[16] Liu Shuqin, Wu Lifang, Gong Yu, et al. Overview of Image Quality Assessment[J]. Sciencepaper Online, 2011, 6(7): 501-506(刘书琴, 毋立芳, 宫玉, 等. 图像质量评价综述[J].中国科技论文在线, 2011, 6(7): 501-506)
[17] Carper W J. The Use of Intensity-Huesaturation Transformations for Merging SPOT Panchromatic and Multispectral Image Data[J]. Photogrammetric Engineering and Remote Sensing, 1990, 56(4): 459-467
[18] Liu Yuqiong. Study on Pixel-Level Multi-source Remote Sensing Data Fusion Technique[D]. Xi’an: Chang’an University, 2007(刘宇琼. 基于像素级多源遥感影像融合技术的研究[D]. 西安:长安大学, 2007)
[19] Mallat S G. A Theory for Multi-resolution Signal Decomposition: The Wavelet Representation[J]. IEEE Transation PAMI, 1989, 11(7): 674-693
[20] Wang Zhengyou, Huang Longhua. Image Quality Assessment Method Based on Contrast Sensitivity[J]. Computer Application, 2006, 26(8): 1 857-1 859(王正友, 黄隆华. 基于对比度敏感度的图像质量评价方法[J]. 计算机应用, 2006, 26(8): 1 857-1 859)
[21] Shen J H. On the Foundations of Vision Modeling I. Weber’s Law and Weberized TV Restoration[J]. Physica D Nonlinear Phenomena, 2003, 175(3): 241-251
[22] Mannos J L, Sakrison D J. The Effects of a Visual Fidelity Criterion on the Encoding of Images[J]. IEEE Trans Information Theory, 1974, 4:525-536
[23] Hu Xuming, Zhang Dengfu, Nan Dong, et al. Method of Image Visual Quality Evaluation Based on Human Visual Characteristics[J]. Computer Application, 2012, 32(7): 1 882- 1 884,1 889(胡许明, 张登福, 南栋,等. 基于人眼视觉特征的图像视觉质量评价方法[J]. 计算机应用, 2012, 32(7):1 882-1 884, 1 889)
[24] Li Wenjuan. Study on the Algorithms for HVS-Based Color Image Measurement[D]. Changchun: Jilin University, 2006(李文娟. 基于视觉特性的彩色图像质量评估研究[D]. 长春:吉林大学, 2006)